containment levels. The guidelines spec- 
ify P2+EK2 levels for such work. There 
was considerable discussion concerning 
the advisability of recommending lower 
containment (P2 EK1) when the DNA 
is isolated from embryonic tissue or 
germ-line cells from cold-blooded verte- 
brates. Those supporting lower contain- 
ment levels argued that the justification 
for P2+ EK2 was the possibility that cold- 
blooded vertebrates may carry viruses 
and that the distinction between adult 
a id germ-cell tissue is real. Others ar- 
gued that, contrary to the situation w'ith 
primate DNA. viruses are not a central 
problem with cold-blooded vertebrates 
and therefore no distinction should be 
made on the basis of tissue origin. Fi- 
nally, the committee recommended, on a 
divided vote (8 to 4*, to adopt P2-EK1 
when the cold-blooded vertebrate DNA 
is isolated from embryonic tissue or 
germ-line cells. Upon reviewing these 
considerations, I have decided to retain 
the containment levels for embryonic or 
germ-line DNA from cold-blooded verte- 
brates as recommended by the commit- 
tee. 
In April the committee also reviewed, 
at our request, the classification of ex- 
periments where DNA is derived from 
other cold-blooded animals or lower eu- 
karyotes. Several commentators, for ex- 
ample, had been concerned about the fact 
that insects are known to carry agents 
pathogenic to man. In the committee re- 
view, it was noted that viruses carried by 
insects and known to transmit disease to 
man are RNA rather than DNA viruses 
and do not reproduce via DNA copied 
from RNA. In order, however, to make 
the intent clearer, the guidelines have 
been rewritten for experiments of this 
class. New language Is inserted to ensure 
that strict containment levels are em- 
ployed when the DNA comes from known 
pathogens or species known to carry 
them. Further, to reduce the potential 
hazards, we have also included in the 
guidelines the requirement that any in- 
sect must be grown under laboratory con- 
ditions for at least 10 generations prior 
to its use as a DNA source. 
2. As alluded to above, certain com- 
mentators expressed concern that when 
E. coli becomes the host of recombinant 
DNA from prokaryotes with which DNA 
Is not usually exchanged, there is hazard 
of altered host characteristics resulting 
from translation of the DNA into func- 
tioning proteins. The committee was 
asked to review the guidelines and take 
Into account this potential hazard. They 
agreed that the containment levels 
should be increased for this category of 
experiment, from P2+EK1 to either 
P2+EK2 or P3-f-EKl. That recommen- 
dation Is included In the present guide- 
lines. 
Comments were made concerning that 
class of experiments in which the recom- 
binant DNA, regardless of source, has 
been cloned. A clone Is a population of 
cells derived from a single cell and there- 
fore all the cells are presumed to be gen- 
etically identical. As outlined in the pro- 
posed guidelines, clones could be used at 
lower containment levels If they had 
NOTICES 
been rigorously characterized and shown 
to be free of harmful genes. Several com- 
mentators inquired how the characteri- 
zation was to be performed and the free- 
dom from harmful genes demonstrated. 
Although the committee acknowledges 
that these terms are unavoidably vague, 
they do cite appropriate scientific meth- 
ods to make relevant determinations. 
Again, this is a rapidly changing area and 
more clarity and precision can be ex- 
pected with experience. Reduced con- 
tainment requirements for this class of 
experiment are warranted because of the 
purified nature of clones. Further, the 
granting agency must approve the clone 
before containment conditions can be re- 
duced, thus providing an additional ele- 
ment of review. 
4. Another comment was related to the 
use of DNA from organelles (intracellu- 
lar-elements that contain special groups 
of genes for particular cell functions'. 
Concern was expressed about the poten- 
tial contamination of purified organelle 
DNA with DNA from viruses because of 
the similarity of their structures. The 
committee agrees, and the guidelines now 
specify a requirement, that the organ- 
elles be isolated prior to extracting DNA 
as a further means of reducing the haz- 
ard of viral contamination. 
5. Some commentators were troubled 
about the lowering of containment for 
that class of experiments involving re- 
combinations with cell DNA segments 
purified by chemical or physical methods. 
They asked that procedures for deter- 
mining the state of purification be more 
fully detailed and that the Recombinant 
Advisory Committee certify the purity. 
There are, however, appropriate tech- 
niques, such as gel electrophoresis, with 
which a purity of 99 percent by mass can 
be achieved and ascertained. There is no 
way for the committee to certify these 
results beyond repeating the experiments 
themselves. These techniques are well 
documented and described in the litera- 
ture. I do not believe it is necessary or 
feasible for the committee to review each 
procedure for purification of DNA. 
6. Comments were made concerning 
the use of DNA derived from animal vi- 
ruses. It was urged that containment lev- 
els for this class of experiment be in- 
creased. On the basis of my review. I find 
the containment conditions appropriate 
to the potential hazard posed. As defined 
in the guidelines, experiments are to be 
done at very strict levels of containment 
and these can be lowered only when the 
cloned DNA recombinants have been 
shown to be free of possibly harmful 
genes by suitable biochemical and bio- 
logical tests. This also pertains to DNA 
that is copied from RNA viruses. In no 
instance are the guidelines more lenient, 
and in most Instances they are more 
stringent than conditions obtaining In 
many laboratories where such viruses are 
studied in non-DNA-recombinant exper- 
iments. 
VI. CLASSIFICATION OF EXPERIMENTS USING 
CONTAINMENT SYSTEMS OTHER THAN E. 
COLI K— 12 
1. No issue with regard to these guide- 
lines raised more comment than the use 
of animal viruses as vectors. Of special 
concern to many commentators was the 
use of the simian (monkey) virus 40 
(hereafter “SV40"). Some suggested a 
complete ban on the use of this virus 
others urged its retention as a vector 
SV40 is not known to produce any di.sea.-e 
in man. although it can be grown in hu- 
man celLs and on very rare occasions ha 
been isolated from humans. Many hu- 
mans have received SV40 virus inadvert- 
ently in vaccines prepared from vim- 
grown in monkey kidney-cell culture- 
An intensive search has been made and 
is continuing for evidence that SY40 
might cause cancer or be otherwise path- 
ogenic for man. At present, it- is my view 
that the extensive knowledge we have ei 
SV40 virus provides us with sufficient so- 
phistication to ensure its safe han- 
dling under the conditions developed for 
its use in the guidelines. 
I believe work with SV40 should con- 
tuiue under the cost careful conditions 
but I do recognize and appreciate the 
concerns expressed over its possible 
harmful effects in humans. In light of 
these concerns, I asked the Recombinant 
Advisory Committee to review this sec- 
tion of the guidelines. The committee 
reconsidered the containment conditions 
for this class of experiments and judged 
them appropriate to meet the potential 
hazards.’ 
This class of experiments will proceed 
under the most careful and stringent 
conditions. Work with SV40 virus will be 
done at the maximum level of physical 
containment <P4). The extraordinary 
precautions required in a P4 facility less- 
en the likelihood uf a potential hazard 
from this work. Only defective SV40 
virus will be used as vector: that is. the 
SV40 virus particles that carry the for- 
eign DNA cannot multiply by themselves 
When a number of strict conditions are 
met, this work will be permitted to go on 
at the third level of containment 'P3 
which in itself requires care and preci- 
sion. It should be noted that SV40 virus 
and its DNA can be efficiently disinfected 
by Clorox and autoclaving. These are 
customary procedures for disinfecting 
glassware and other items used in SV40 
animal-cell work. 
Some commentators suggested that 
the containment criteria for experiments 
using polyoma virus as the vector be 
strengthened. There is no evidence that 
polyoma infects humans or replicates to 
any significant extent in human cells. It 
holds promise as a vector, as is more fully- 
documented In an appendix to these 
guidelines. 
2. Several commentators found the 
guidelines inadequate regarding experi- 
ments with plant host-vector systems. 
Because NIH shared these concerns, a 
group with extensive experience with 
plants was appointed to review this sec- 
tion. The group met concurrently with 
* One member dissented from tins position. 
During the discussion, additional language 
was recommended (and adopted) to ensure 
that the defective SV40-virus 'helper-virus 
system, with Its Inserted non-SV40 DNA seg- 
ment, does not replicate in human cells with 
significantly more efficiency than does SV4 o. 
FEDERAL REGISTER, VOL. 41. NO. 131 — WEDNESDAY, JULY 7, 1976 
